Hi Philip,

On 13 Nov 2026, at 16:03, Philip Homburg <[email protected]> wrote:

> The tension between draft-ietf-dnsop-multi-alg-rules and
> draft-jabley-dnsop-local-signing-algorithm-policy got me thinking.

I don't think there is actually any fundamental tension between these two 
drafts (or between local-signing-algorithm-policy and RFC 4035). In fact, I 
think they are complimentary -- my draft describes how a local policy that is 
not accommodated by 4035 or multi-alg-rules is still possible, and how clients 
affected by that local policy can be informed about it. The provision for local 
policy to override the usual validation approach is mentioned explicitly in RFC 
4035 and I don't see that multi-alg-rules proposes to remove it.

When it comes to the PQ downgrade problem itself and not just my draft's 
approach to implementing protection or its use of local policy to implement it, 
I see what you are suggesting. I wonder slightly whether one bit is enough 
(maybe there are more future scenarios than PQ-safe and PQ-unsafe), but only 
slightly. Your idea seems clever.

I don't think these ideas are in competition, though. We have substantially no 
experience deploying PQ-safe algorithms in DNSSEC, and we also have no 
experience with actual quantum computers breaking PQ-unsafe algorithm (or not 
being able to break those that are declared to be PQ-safe). At any time, 
different resolvers might assess the underlying risks differently. It seems to 
me that having a worked example of how to manage a particular local validation 
policy will continue to be useful, regardless of what other clever ideas exist.


Joe
_______________________________________________
DNSOP mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to